[
https://issues.apache.org/jira/browse/FLINK-3654?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15228020#comment-15228020
]
ASF GitHub Bot commented on FLINK-3654:
---------------------------------------
Github user aljoscha commented on the pull request:
https://github.com/apache/flink/pull/1850#issuecomment-206259647
I think we have a winner now. I changed the backup options to not back up
the WAL. This seems to have been the problem. Only logical since there is no
WAL anymore...
Before, 2 out of 5 travis runs (i.e. 5*5) would have errors. Now all 5 runs
that I started were successful.
> Disable Write-Ahead-Log in RocksDB State
> ----------------------------------------
>
> Key: FLINK-3654
> URL: https://issues.apache.org/jira/browse/FLINK-3654
> Project: Flink
> Issue Type: Improvement
> Components: Streaming
> Reporter: Aljoscha Krettek
> Assignee: Aljoscha Krettek
>
> We do our own checkpointing of the RocksDB database so the WAL is useless to
> us. Disabling writes to the WAL should give us a very large performance boost.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
